In a milestone achievement for quantum technology, researchers have demonstrated an unmistakable quantum advantage using entangled light on a scalable photonic platform, as recently published in the prestigious journal Science. This breakthrough affirms that quantum systems designed with entanglement can drastically reduce the experimental efforts required to characterize complex, noisy environments—an advance that promises to propel quantum sensing and machine learning to new heights.
The technical team, led by Ulrik Lund Andersen at the Technical University of Denmark (DTU), showcased for the first time how entangled light can be harnessed to learn the intrinsic noise properties of a quantum system with exponentially fewer measurements compared to any classical strategy. The implications are profound: what would take classical methods nearly 20 million years to achieve was completed in just 15 minutes using a carefully engineered entangled photonic setup.
At the core of the experiment lies an optical parametric oscillator (OPO), often referred to as a “squeezer,” which manipulates quantum fluctuations of light through a nonlinear crystal inside an optical cavity. This device generates entangled light beams that are quantum correlated in such a way that measurements of one beam instantaneously reveal information about the other, enabling a joint measurement strategy that uncovers system noise far more efficiently than classical probing.
Noise characterization in quantum systems is notoriously challenging because quantum noise itself forms part of the measurement signal. As system complexity grows, the number of required measurements typically increases exponentially, creating a barrier to practical analysis and calibration. By exploiting quantum entanglement, the DTU group succeeded in bypassing this obstacle, proving that quantum correlations can be utilized to circumvent classical limitations in learning system behavior.
The experimental apparatus operated at telecom wavelengths utilizing standard optical components, an intentional design choice to demonstrate robustness against realistic losses and imperfections. This practical approach highlights that the observed learning advantage originates fundamentally from the entangled measurement process itself, rather than dependence on an idealized, lossless environment or perfect detectors.
Two beams produced by the squeezer were allocated asymmetrically: one acted as a probe interacting with the noisy system, while the other served as a stable reference. The configuration allowed simultaneous joint measurements, where the comparison between probe and reference beams largely canceled out the detrimental effects of measurement noise, thereby extracting maximal information per trial and dramatically reducing the total number of experiments necessary.
This landmark demonstration validates theoretical predictions outlined earlier in the field, including a notable 2024 study on entanglement-enabled learning advantages for bosonic channels. Their prior theoretical groundwork laid the foundation for this empirical realization, showcasing the direct link between entanglement and enhanced information gain in quantum systems.
While the current work focused on a simplified, controlled optical channel with a fixed noise pattern, the researchers emphasize that the methodology is widely applicable to a plethora of quantum systems exhibiting noise correlations. This universal versatility points toward future applications in quantum sensing devices, quantum communications, and quantum-enhanced machine learning platforms, where rapid and accurate noise characterization is essential.
Jonas Schou Neergaard-Nielsen, co-principal investigator and associate professor at DTU Physics, remarked on the significance of the results by underscoring that, unlike many theoretical quantum proposals that await practical demonstration, their experiment unmistakably accomplishes what no classical mechanism can replicate. This experimental proof of quantum superiority marks a turning point, confirming that quantum strategies will soon begin to practically outperform classical counterparts.
The collaborative research effort extended beyond DTU to include leading institutions such as the University of Chicago, Perimeter Institute, University of Waterloo, Caltech, MIT, and KAIST, reflecting the global nature of the quest to unlock quantum technologies’ full potential. Together, the team combined theoretical expertise and cutting-edge experimental skill to realize a scalable optical platform poised to transform quantum measurement science.
This breakthrough extends beyond pure academic interest, as the researchers anticipate that their approach will inspire immediate advances in quantum-enhanced metrology and sensing. The demonstrated efficiency gain will likely accelerate experimental throughput in diverse quantum systems, facilitating real-world implementation of quantum algorithms for noise reduction, state discrimination, and beyond.
Moreover, by employing well-understood photonic components and operating in the telecom band, the setup aligns naturally with existing fiber-optic infrastructure, ensuring that integration into current optical communication and quantum network technologies is feasible. This compatibility promises rapid technology transfer from laboratory demonstrations to commercial quantum devices.
In summary, this work sets a new standard, dissecting the quantum-classical boundary and exhibiting a clear quantum advantage in learning and characterizing complex noisy systems with unprecedented efficiency. It marks a decisive moment in quantum science, showing not only that quantum entanglement can fundamentally accelerate information acquisition but also that such advantages are accessible with realistic, scalable photonic technologies.
Subject of Research: Quantum advantage in noise characterization using entangled light on a scalable photonic platform
Article Title: Quantum learning advantage on a scalable photonic platform
News Publication Date: 25-Sep-2025
Web References:
https://science.org/doi/10.1126/science.adv2560
https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.133.230604
Image Credits: Photo by Jonas Schou Neergaard-Nielsen (DTU Physics)
Keywords
Quantum advantage, entanglement, photonic platform, noise characterization, optical parametric oscillator, squeezed light, quantum sensing, quantum metrology, scalable quantum systems, telecom wavelength, quantum machine learning, quantum noise
